Characterization of new formulations for the rotational molding based on ethylene-propylene copolymer/graphite nanocomposites.

INTRODUCTION

Among the traditional polymer processing techniques, the rotational molding allows the processing of large parts, in a single step. Nevertheless, the main disadvantage of rotational molding is the long times of processing cycles, which limits its uses. Polyethylene is one of the polymers that is most commonly used as a matrix for rotomolding. However, polyolefin matrices formulated with nanofillers have not been tested for rotomolding. Thus, the aim of this study is to develop new materials having good thermal properties. That is why thermally conductive fillers were incorporated in order to reduce the heating time of the polymer powder. Blends of polymer with conducting fillers including CB or metal powder have been investigated in the past few decades. But increasing the thermal conductivity of polyolefins remains a major challenge.

An original way is to incorporate graphite, that is a filler with enhanced thermal conductivity (134 W/mK) and a lamellar shape showing a high aspect ratio (100). But to take benefit from its properties, a uniform and fine dispersion all over the scales of graphite is the key point to succeed. Recently, polymer/graphite nanocomposites have raised some interest because of their potential conductive properties. Various approaches for making polymer/graphite blends have been developed and some works on the preparation of conductive expanded graphite (EG)/polymer nanocomposites were reported in the literature [1-4]. Zheng et al. [5] successfully prepared polymethyl methacrylate (PMMA)/EG nanocomposites by direct solution blending of PMMA with EG. According to them, the improvement of the conductivity of the polymer/EG nanocomposites can be attributed to the formation of a conducting network of graphite sheets within the polymer matrix. Chen et al. [6] reported that the nanoscale dispersion of graphite flakes in the maleated polypropylene matrix could be achieved by using an intercalation method. Liu et al. [7] compared the conductivity of polystyrene/EG nanocomposites prepared by melt blending and in situ polymerization technique. Krupa et al. [8] studied the thermal properties of HDPE/EG and PS/EG. For both matrices, the thermal conductivities and diffusivities increased nonlinearly with an increasing percentage of graphite, but this evolution did not show a percolation ratio. The study of thermal properties of nylon 6/EG [9] also showed an increase of the thermal conductivity with an increasing percentage of filler. Moreover the mechanical properties of these compounds [10] showed a considerably higher modulus than one of composites made with commercially available carbon reinforcing materials [i.e. vapor grown carbon fibers (VGCF) and carbon black (CB)]. The purpose of all these researchers is to obtain the composites with graphite particles dispersed uniformly in a continuous polymer matrix. Generally, it is difficult to be fulfilled due to their incompatibility. Thus, improvement of the incompatibility between polymer and graphite is important.

In this work, the selected matrix is an ethylene-propylene copolymer, judiciously chosen to improve the impact resistance of rotomolded parts. The main objective of this article is to determine the role of the interactions on the dispersion of fillers in the matrix; thus we have studied the effect of both parameters: (i) the surface treatment of the fillers and (ii) incorporation of a coupling agent in the formulation to modulate these interactions.

EXPERIMENTAL

Materials

Ethylene-propylene copolymer (EP) BF330MO from Borealis was used as matrix. And maleated EP copolymer (EP-g-MA) which trade name is Fusabond MD353D from Dupont was chosen as compatibilizer. The main characteristics of both materials are reported in Table 1.

The graphite used in this study was synthetic flake graphite with a size included between I and 10 [micro]m, supplied by Timcal with the reference SFG6.

Surface Treatments of Graphite

Acid Treatment. Intercalation of graphite was performed by immersion of graphite flakes in a mixture of a concentrated nitric acid and sulphuric acid for one day [11, 12].

n(graphite) + n [H.sub.2]S[O.sub.4] + n/2(o) [right arrow] n(graphite.HS[O.sub.4]) + n/2 [H.sub.2]O (1)

(o), Oxidizing agent: here nitric acid; (graphite.HSO4), intercalated graphite (GIC).

The resulting material, denoted as graphite intercalation compound (GIC), is the stacking of the successive layers including carbon layers and intercalated layers. But the real situation is frequently an irregular intercalation of layers [13]. Then the graphite is washed with water until this one becomes neutral and is filtered; the graphite is dried at 100[degrees]C in order to remove water. Finally, the particles are treated at 1050[degrees]C during 15 s. Fast heating of intercalated graphite flakes at a high-enough temperature leads to exfoliation. Thus formed vermicular graphite is also known as EG. This treatment leads to the creation of polar groups on the edges of the graphite sheets.

Treatment With Stearic Acid of Untreated and EG [14]. Graphite (5 g), toluene (150 mL), and stearic acid (C[H.sub.3](C[H.sub.2])[.sub.16]COOH-60 mg) were mixed during 8 h with a centrifugal spreader, and then the fillers were filtered and dried. This treatment is commonly used for the compounds polypropylene/calcium carbonate in order to improve the filler/matrix interactions by making fillers more organophilic.

Processing of Nanocomposites

EP copolymer/lamellar graphite nanocomposites were processed by melt mixing in a twin-screw extruder corotating microextruder from DSM at 210[degrees]C for 5 min at a screw speed of 60 rpms [15]. The formulations are listed in the Table 2.

Characterization of Graphite Powders

The characterization of the crystalline structure of graphite has been performed by wide-angle X-ray diffraction (WAXD). The instrument was a diffractometer Siemens D500 with a goniometer Bragg-Brentano; a copper cathode was used as X-rays source ([lambda] = 0.154 nm). The powders were compacted with a pressure of 375 MPa and were deposited on a glass substrate. The measurements were carried out between 1[degrees] and 80[degrees] 2[theta] by step of 0.02[degrees] at room temperature. The size of diffracting object L, which is in this case the primary particle, can be determined with the Scherrer equation.

L = (0.9 x [lambda])/([DELTA](2[theta]) x cos ([theta])) (2)

With L, size of diffracting object; [lambda], wavelength of copper; [DELTA](2[theta]), width of peak at middle height; [theta], angle of diffraction peak.

The surface treatment performed on graphite was put into evidence by surface energy measurements. The powders were compacted with a pressure of 1250 MPa under vacuum to realize some pellets. The contact angles between a liquid test and the powder solid surface were measured on pellets at room temperature using the sessile drop method with Digidrop from GBX (France). The distilled water, a polar liquid and diiodomethane, a dispersive liquid were used like test liquid. The relationships between the contact angle and the surface energy was obtained by applying the Owens-Wendt theory.

Characterization of Nanocomposites

Filler Dispersion Characterization. Evidences of graphite sheet dispersion within the matrix were provided: at the nanoscale by WAXD between 1[degrees] and 80[degrees] 2[theta] by step of 0.02[degrees] at room temperature by using the Scherrer equation, and at a larger scale, i.e. up to few micrometers squares, by scanning electron microscopy. It was an environmental microscope with a variable pressure HITACHI S3500N. The samples were fractured in the liquid nitrogen and were observed without being metallized. To characterize the interactions between fillers and matrix in processed nanocomposites, the rheological behavior of these materials was studied as well. An increase of viscosity was significant of the intimate contact between fillers and matrix, i.e. a succeeded dispersion. The dynamical rheometer RDA II supplied by Rheometric was allowed to follow the evolution of complex viscosity and mechanical modulus as a function of time, temperature, and strain. In our case, a plan/plan geometry has been used. The analysis has been performed at the temperature of the rotational molding (230[degrees]C) with a strain of 5% and a shear rate of 1 rad [s.sup.-1] in order to simulate the rotational molding process.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

Characterization of Compound Final Properties

The rheological analysis allowed also to check if the viscosities of the filled copolymers fit the viscosity range suitable for rotational molding.

Thermal conductivities were investigated on compression moulded nanocomposite samples. The measurements were performed using a homemade apparatus developed by CETHIL, Insa-Lyon. The sample was set between a cold and a heated source. The test consisted keeping a constant heat flux. From the value of the heat flux emitted by the heated source and the measurement of temperature difference between the two sources, the thermal conductivity of the sample could be determined.

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

RESULTS AND DISCUSSIONS

Characterization of the Surface Treatments on Lamellar Graphite

The surface energy measurements show that the graphite has a surface energy much higher than the one of the polyolefin matrix (Table 3), which does not allow the wetting of the filler by the matrix and does not favor the interactions. This high surface energy is due to the presence of the polar groups already existing on the sheets of graphite that are all the more revealed by the acid treatment.

Graphite is a layered material, which shows a multi-scale structure based on primary particles themselves made of graphitic sheets themselves composed of stacked carbon layers as drawn in Fig. 1. Graphite sheets associated between themselves with weak Van Der Waals interactions may move and swell after intercalation organic chains. However, the d-spacing between carbon layers is always kept at 3.35 [Angstrom] (Fig. 2). Two additional peaks at 2[theta] = 54.8[degrees] and 87[degrees] (associated to distances 1.67 and 1.12 [Angstrom], respectively) frequently observed on X-ray spectra characterize the regular crystalline organization of the carbon layers (Fig. 3). After acid treatment, the Van Der Waals forces existing between sheets allow acid molecules to intercalate into the interplanar spaces of the graphite sheets. The interplanar spacing, also called galleries, are thus increased, consequently on the treated graphite X-ray spectra (treated with acid, with stearic acid, or both), an additional peak at 2[theta] = 13.5[degrees] i.e. d = 6,55 [Angstrom], is observed, significant of sulfuric acid intercalation (Fig. 4). By referring to the literature, the distances d = 7.55 [Angstrom] [13], d = 9.81 [Angstrom] [2] are characteristic of an EG whereas the distance d = 6.59 [Angstrom] [4] is significant of an intercalated graphite. But commonly these peaks obtained at small angles indicate a successive stacking of carbon layers and [H.sub.2]S[O.sub.4] layers. With the Scherrer's equation, the size of the diffracting object, i.e. the primary particle, can be determined (Table 4). The size of primary particle can be divided by a factor 2 by treating graphite with acid and stearic acid. In all the cases, the surface treatments induce an expansion of graphite.

Analysis of EP/Graphite Nanocomposite

Surface Properties. When the matrix is compatiblized (Fig. 5), the adding of lamellar graphite modifies the surface properties of copolymer matrix. For most of the filled blends, an increase of polar component is observed due to the presence of polar groups, initially existing on untreated graphite and all the more revealed by the acid treatment. But oppositely, the modification with stearic acid leads to an increase of dispersive component and a strong decrease of polar component. The surface modification with alkyl chains of stearic acid makes more dispersive, i.e. hydrophobic the graphite surface. We can hope that interactions between fillers and matrix are better since surface energies are almost similar. The Fig. 6 gives a schematic drawing of interactions involved in both treatment, i.e. acid and stearic acid. On the other hand, for the blends performed with two times modified graphite (acid + stearic acid), it would seem that the effect of the two treatments has been annihilated since dispersive and nondispersive components are similar to pure matrix one. The same observations were made on uncompatibilized matrix.

Rheological behavior

1. Effect of graphite ratio: The rheological measurements show clearly an increase of viscosity with the increasing ratio of graphite (Fig. 7). With 5% of graphite, the viscosity remains similar to the pure polymer one: when the graphite sheets are well dispersed, the lamellar graphite has a lubricant effect on EP matrix. From 10%, the adding of graphite modifies the rheology of the pure matrix, the viscosities are much higher than pure polymer one. On the other side, the slow decrease of matrix viscosity attributed to the copolymer degradation is not observed on the nanocomposites. This effect can be attributed to the graphite that balances this viscosity fall by the "nanocomposite effect." At low shear rate, the viscosity increase with the analysis time is due to the structuration of fillers within the matrix that takes place under the effect of the diffusion of the polymer chains.

2. Effect of graphite surface treatment: The rheological measurements reported in Fig. 8 show that the viscosity increases during the rotomolding process simulated by an analysis time of 1800 sec at 230[degrees]C since the structuration of graphite sheets takes place. The initial viscosity means the dispersion level of graphite into the matrix. The both treatments performed separately with acid or stearic acid, are efficient and favor the dispersion since the highest viscosity is obtained for the compounds based on acid or stearic acid treated graphite. On the other hand, the double treatment has not absolutely a synergetic effect since the viscosity level is below one of the blend performed with untreated graphite. Maybe successive treatments do not contribute to a good dispersion, in terms of granulometry of the fillers before blending.

[FIGURE 6 OMITTED]

[FIGURE 7 OMITTED]

3. Effect of the compatibilizer: The addition of a maleated EP copolymer has. for consequence, to decrease the viscosity of matrix because of its low molar mass like shown in Fig. 9. Afterwards, the same tendency is observed for the filled compounds. The acid treatment and the treatment with stearic acid performed separately on graphite have the same effect on the compatibilized matrix as on the pure copolymer. Consequently, these both surface treatments realized separately on graphite sheets improve the graphite dispersion in the compatibilized matrix as well.

To conclude, the viscosities measured on nanocomposites are indeed higher than one of virgin polymer but remain suitable for rotational molding process [16]: the complex viscosity of the compounds remains lower than 3000 Pa s.

[FIGURE 8 OMITTED]

[FIGURE 9 OMITTED]

Crystalline Microstructure 

The incorporation of lamellar graphite within a semi-crystalline matrix can modify the crystalline microstructure of resulting nanocomposite. Can the high graphite surface have its effect on the matrix crystallization? The XRD analysis performed on all the compounds (reported in Fig. 10 and analyzed in Table 5), reveals peaks typical of virgin matrix (9.5[degrees], 14.2[degrees], 16.9[degrees], 18.6[degrees], 21.2[degrees], 21.8[degrees], 25.5[degrees], 28.6[degrees], 43[degrees]) and graphite (26.6[degrees], 54.8[degrees], 87[degrees]). We can observe that the intercalation of polymer has not modified the space between carbon layers since the peak situated at 2[theta] = 26.6[degrees] for d = 3.35 [Angstrom] is always present. On the other side the peak at 2[theta] = 13.5[degrees] significant of acid intercalation observed on the treated graphite spectra (Fig. 4) disappeared, reflecting the extended dispersion during the nanocomposite processing step. The only modification observed on the crystalline microstructure is the appearance of a peak at 2[theta] = 16.2[degrees] on the compound EP(15%)-st put into evidence on zoom reported in Fig. 11. According to the literature [3], this peak is assigned to [beta] crystalline form of polypropylene that mainly crystallizes under the [alpha] form. The whole of peaks is summarized in Table 6. The crystallization under the [alpha] form is usually obtained in presence of a nucleating agent, which can be is assimilated here to the stearic acid modified graphite.

[FIGURE 10 OMITTED]

From the half height width of X-ray peaks, we can determine the size of primary particles (given in Table 7) and check the intercalation of polymer chains between the graphite sheets, that is a key mechanism for an exfoliation of graphite sheets. At the primary particle scale, we can conclude that the addition of the compatibilizer is very helpful to exfoliate the primary particles since a size decrease is observed. In the same way, the treatment alone with stearic acid and the double treatment with acid and stearic acid have a positive effect on the particles delamination. The primary particles are thinner when the matrix is compatibilized or the graphite is treated with stearic acid or both (acid + stearic acid).

[FIGURE 11 OMITTED]

Morphological Analysis

At a higher scale, microscopic analysis by MEB (shown in Figs. 12 and 13) allow to check if the wetting of graphite sheets by polymer took place.

For the matrix uncompatibilized, the graphite's sheets appear bright on micrography and are well observable in the pictures from a to d, which indicates that the surface treatment alone is not enough to favor the filler/matrix interactions and to lead to a homogeneous dispersion. In the same way, the compatibilizer alone is not sufficient to help the graphite dispersion (Fig. 13a). The stearic treatment does not seem efficient at all (Fig. 13c). The morphological analysis does not corroborate the viscosity measurements, which showed a better dispersion for the two similar formulations. Furthermore, the association of the compatibilizer with double treatment (acid + stearic acid) or the acid treatment alone seems to be the best optimized formulation to improve the interactions filler/matrix: a good wetting of graphite sheets by polymer matrix is clearly put into evidence in Fig. 13b and d. However, the evaluation of organic fillers dispersion in organic matrix remains difficult by electronic microscopy.

[FIGURE 12 OMITTED]

[FIGURE 13 OMITTED]

Thermal Properties

To check if these materials could be used for the rotational molding, measurements of thermal conductivities have been made. In Fig. 14, a nonlinear increase of thermal conductivity is observed with an increase of graphite content. Thus, the heating time of polymer powder should be improved.

[FIGURE 14 OMITTED]

Some measurements of electrical conductivities are in progress in order to determine the percolation ratio and to make a correlation with thermal conductivity measurements [8]. But for a graphite weight percentage of 15%, the nanocomposite was not always an electrical conductor. In fact, the increase of electrical conductivity was observed for a content of graphite higher than 20%.

CONCLUSION

This study has allowed to process EP copolymer/lamellar graphite compounds, which can be used for rotational molding. The formulation based on EP copolymer/maleated EP copolymer/SFG6 graphite treated with acid is the most suitable formulation in terms of dispersion and viscosity for rotational molding. The interactions between fillers and matrix are improved through hydrophilic interactions established between acid-treated graphite and grafted copolymer, this last one being bonded with the matrix. With the optimized formulation (EP(15%)-a), some rotomolded parts were processed and the heating time of polymer powder has been improved of ~25%.

REFERENCES

1. Y.X. Pan, Z.Z. Yu, Y.C. Ou, and G.H. Hu, J. Polym. Sci. Part B: Polym. Phys., 38, 1626 (2000).

2. W.P. Wang and C.Y. Pan, Polymer, 45, 3987 (2004).

3. T.G. Gopakumar and D.J.Y.S. Page, Polym. Eng. Sci., 44, 6 (2004).

4. M. Uhl Fawn, E. Hiroyoshi, E. Manias, and A. Wilkie, Polym. Degrad. Stab., 89, 1 (2005).

5. W. Zheng, S.C. Wong, and H.J. Sue, Polymer, 43, 25 (2002).

6. X.M. Chen, J.W. Shen, and W.Y. Huang, J. Mater. Sci., 21, 3 (2002).

7. P. Liu, K. Gong, P. Xiao, and M. Xiao, J. Mater. Chem., 10, 933 (2000).

8. I. Krupa and I. Chodak, Eur. Polym. J., 37, 2159 (2001).

9. H. Fukushima, L.T. Drzal, B.P. Rook, and M.J. Rich, J. Therm. Anal. Cal., 85, 1 (2006).

10. H. Fukushima, L. Sung Ho, and L.T. Drzal, ANTEC, 2, 1441 (2004).

11. G.H. Chen, D.J. Wu, W.G. Weng, and W.L. Yan, Polym. Eng. Sci., 41, 12 (2001).

12. G.H. Chen, D.J. Wu, W.G. Weng, and W.L. Yan, J. Appl. Polym. Sci., 82, 2506 (2001).

13. X.S. Du, M. Xiao, Y.Z. Meng, and A.S. Hay, Polym. Adv. Technol., 15, 320 (2004).

14. Y.W. Lee, Polym. Comp., 24, 1 (2003).

15. K. Kalaitzidou, ANTEC, 2, 1533 (2004).

16. R.J. Crawford, Polym. Eng. Sci., 36, 7 (1996).

Emilie Planes, Jannick Duchet, Abdherrahim Maazouz, Jean-Francois Gerard

Universite de Lyon-Laboratoire des Materiaux Macromoleculaires, UMR IMP CNRS #5223, INSA-Lyon, Bat. Jules Verne, 17 avenue Jean Capelle, F-69621 France

Correspondence to: Emilie Planes; e-mail: [e-mail]

TABLE 1. Main characteristics of EP and EP-g-MA.

         % (mass) Ethylene  [bar.Mn] (g/mol)  [bar.Mw] (g/mol)

EP       10                 48,000            278,400
EP-g-MA   -                 11,063             88,500

         [l.sub.p]  MFI (g/10 min)        Grafting ratio (%)

EP       5.8        18.0 @ 230[degrees]C  -
EP-g-MA  8.0        21.3 @ 160[degrees]C  1.4

TABLE 2. Description of studied formulations (the ratio of graphite is
expressed in weight).

EP                       95           90            85

EP-g-MA                  0             0             0
Graphite SFG6            5            10            15
Without treatment        EP(5%)-unt   EP(5%)-unt    EP(5%)-unt
Acid treated             EP(5%)-a     EP(10%)-a     EP(15%)-a
Stearic acid treated     EP(5%)-st    EP(10%)-st    EP(15%)-st
Double treatment with    EP(5%)-a-st  EP(10%)-a-st  EP(15%)-a-st
  acid and stearic acid

EP                       80.75

EP-g-MA                   4.25
Graphite SFG6            15
Without treatment        EPg(5%)-unt
Acid treated             EPg(15%)-a
Stearic acid treated     EPg(5%)-st
Double treatment with    EPg(15%)-a-st
  acid and stearic acid

TABLE 3. Surface energy of different graphite powders compared to matrix
one.

                                 Surface  Nondispersive  Dispersive
                                 energy   component      component
                                 (mN/m)   (mN/m)         (mN/m)

Graphite SFG6 untreated          58.2     20.7           37.5
Graphite SFG6 treated with acid  61.6     22.6           39.0
EP                               27.6      3.0           24.6

TABLE 4. Size of the primary panicles after different surface
treatments.

                                                             Graphite
                                                             SFG6
                                                             treated
                     Graphite   Graphite      Graphite SFG6  with acid
                     SFG6       SFG6 treated  treated with   and stearic
                     untreated  with acid     stearic acid   acid

Size of diffracting  523        392           304            237
  object (primary
  particle)
  ([Angstrom])

               Total surface energy (mN/m) [gamma] = [[gamma].sup.rd] +
               [[gamma].sup.d]
               Non dispersive component  Dispersive component
               [[gamma].sup.nd](%)       [[gamma].sup.d](%)

EPg(15%)-a-st  10%                       90%
EPg(15%)-st     4%                       96%
EPg(15%)-a     32%                       68%
EPg(15%)-unt   20%                       80%
EPg            15%                       85%

FIG. 5. Determination of the dispersive and non dispersive components of
filled and compatiblized compounds. The contribution of the surface
energy measured are expressed in weigh rate.

Note: Table made from bar graph.

TABLE 5. Characteristics of the numeroted peaks in Fig. 8.

Peak  2[theta] ([degrees])  D ([Angstrom])  Peak assigned to

 1     9.5                  9.30            Matrix
 2    14.2                  6.23            Matrix
 3    16.9                  5.24            Matrix
 4    18.6                  4.76            Matrix
 5    21.2                  4.19            Matrix
 6    21.8                  4.07            Matrix
 7    25.5                  3.49            Matrix
 8    26.6                  3.35            Graphite
 9    28.6                  3.12            Matrix
10    43                    2.1             Matrix
11    54.8                  1.67            Graphite
12    87                    1.12            Graphite

TABLE 6. Analysis of peaks ([degrees]2[theta]) significant of the
[alpha]-form and [beta]-form of the polypropylene.

Peak  [alpha]-form   [beta]-form

 2    14[degrees]
13                   16[degrees]
 3    17[degrees]
 4    18.5[degrees]
 5    21[degrees]    21[degrees]
 6    22[degrees]    22[degrees]

TABLE 7. Size of primary particles determined on different nanocomposite
formulations with or without compatibilizer, as a function of graphite
treatment.

                                             Size of primary particle
                                             ([Angstrom])

Graphite untreated                           523
  EP(15%)-unt                                327
  EPg(15%)-unt                               184
Graphite treated with acid                   392
  EP(15%)-a                                  327
  EPg(15%)-a                                 185
Graphite treated with stearic acid           304
  EP(15%)-st                                 184
  EPg(15%)-st                                209
Graphite treated with acid and stearic acid  237
  EP(15%)-a-st                               177
  EPg(15%)-a-st                              193